Although I don't agree with much of this analysis, it is right that solar will be incredibly cheap, and thought-provoking.
(I think the fundamentals of applying learning curve analysis to a LCOE are slightly dodgy, and 30% is too high a rate even for capex).
(I think the fundamentals of applying learning curve analysis to a LCOE are slightly dodgy, and 30% is too high a rate even for capex).
Specifically, I'm a little concerned that a lot of the falls in cost of capital, O&M cost and non-module cost over the last decade have been one-offs. Cost of capital has dropped worldwide for everything, for example.
Our definition of a normal utility-scale PV project has gone from 1MW to at least 50MW, and there aren't that many places where you can just whack down a GW plant, so typical system sizes will cap out. Cleaning panels and mowing vegetation is more and more of the LCOE.
That said, I am one of the experts who has consistently underestimated cost reductions and short-term growth in solar over this decade. (There is an entire chapter in my book about how terrible I am at prediction).
There are, genuinely, advances being made which will continue to reduce capex and even increase capacity factor (the move to bifacial modules adds 4-9% extra generation, which we don't yet include in BNEF LCOE and probably should).
I also can't agree more with the bit titled "Solar is Amazing. It isn’t a Panacea". Any energy forecast going out to, say, 2025 should include solar as one of the cheapest sources of electricity, but seasonality is a huge barrier to high penetration in the electricity mix.
